Glioblastoma, the most aggressive brain cancer, invariably reoccurs after surgery and rapidly develops resistance to radiation therapy and chemotherapy. The invasive nature of glioblastoma is a major cause of therapeutic failure. Furthermore, the study of glioblastoma invasion is particularly challenging due to the lack of good experimental models that recapitulate the tumor microenvironment. For genetic characterization of tumors, most studies have investigated clonal (high-frequency) mutations and have not analyzed sub-clonal (low- frequency) mutations due to the high error rates (10-2 to 10-3) of DNA sequencing methods. Therefore, little is known about genetic variations and heterogeneities that can determine the evolution of invasive and therapy- resistant subpopulations of glioblastoma cells. To recapitulate in vivo 3D tumor microenvironments, we have established nanotopographically defined extracellular matrix (ECM)-mimetic culture platforms. To accurately detect subclonal mutations as well as clonal mutations, we have employed Duplex Sequencing (10-8 to 5x10-8), which is >10,000-fold more accurate than other high-throughput sequencing methods. The overall goal of this project is to characterize genetic variations underlying the invasiveness and radiation resistance of glioblastoma cells in biomimetic culture. We will pursue the following aims: 1) examine the migration properties of patient- derived orthotopic glioblastoma xenograft (GBM/PDX) and radiation-resistant GBM/PDX (r-GBM/PDX) cells in a nanotopographical ECM-mimetic culture under hypoxia. Cell migration property will be monitored as an indicator of invasion via a high-throughput live-cell microscopy; and 2) determine how the mutation load or mutation types that change during the development of radiation resistance correlate with the migration properties of GBM/PDX and r-GBM/PDX cells. We will identify clonally expanding subclonal mutations and de novo mutations that arise during radiation resistance using Duplex Sequencing. This project has implications in the preclinical screening of anti-migratory chemotherapeutic agents, the molecular classification of glioblastoma, and the identification of biomarkers for the early detection of therapy resistance and tumor recurrence.